The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructions for the agent to utilize the rg (ripgrep) utility to scan the local codebase for specific patterns, such as as casts, during the audit flow defined in SKILL.md and references/cast-cleanup.md.\n- [SAFE]: The skill's data access is limited to reading and analyzing local source code files to report on type safety invariants. This behavior is restricted to the local environment and is necessary for its auditing functionality.\n- [PROMPT_INJECTION]: The skill analyzes local source code, which serves as an untrusted data source and presents a surface for indirect prompt injection.\n
Ingestion points: Local files and packages specified by the user during the audit process as described in the SKILL.md entry points.\n
Boundary markers: Absent. The skill does not provide instructions to the agent to ignore or delimit embedded natural language prompts within the source files.\n
Capability inventory: The skill uses file-read tools and shell command execution (rg) to extract and summarize code content (SKILL.md).\n
Sanitization: None. The skill processes the raw content of the audited files without sanitization or validation.

type-safety